Papers with Code 아카이브

3D Face Modelling

태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.

하위 태스크
2
평가표
2
보존된 평가 기록
3

데이터셋

출처에 같은 태스크 이름으로 연결된 데이터셋입니다. 동명 태스크가 포함될 수 있으므로, 비교할 때는 평가 결과의 조건도 확인하세요.

SiW-Enroll

SiW (Spoofing in the Wild) is a face anti-spoofing dataset recently introduced in [29] where images are extracted from short videos captured at high resolution and 30 frames per second. In total, 4,478 videos are collected from 165 subjects including variations in spoof type, recording device, illumination condition, pose and facial expression. To define train and test sets, we start by following the splitting system described as Protocol 1 in [29]. We further subsample train and test sets by extracting 1 every 10 frames (or every 0.33 seconds) in each video, since consecutive frames are almost identical. Finally, since we observe that simple models can reach very high accuracy, we further make the task harder by creating 2 separate training and evaluation folds in a way that the model is trained and tested on different spoof types. Since the original dataset is collected using two different capturing devices, we can use this information for the definition of enrollment sets. Indeed, in a real scenario, the subject would have to enroll twice when using two different devices, meaning that a unique enrollment set will be generated for each combination of subject and capturing devices. Capturing the sensor bias in the enrollment can be particularly beneficial, for example to detect if the resolution of a face changes, which can indicate a replay attack. To extract enrollment data we choose a single video from each subject and sensor among the available ones. We arbitrarily pick the video without illumination changes and with variation in subject pose, as it resembles the enrollment conditions required in real devices. We then equidistantly sample N frames over the video to construct the enrollment set. As shown in Fig. 2, this allows capturing different poses and facial expressions from the subject. Finally, we exclude all the frames in this video from training and test data to avoid in341 formation leakage and match all the remaining queries from the same subject and sensor to their unique enrollment set. (See paper for additional details)

UFPR-Periocular

The UFPR-Periocular dataset has 16,830 images of both eyes (33,660 cropped images of each eye) from 1,122 subjects (2,244 classes). All the images were captured by the participant using their own smartphone through a mobile application (app) developed by the authors. There are 15 samples from each subject's eye, obtained in 3 sessions (5 images per session) with a minimum interval of 8 hours between the sessions. The images were collected from June 2019 to January 2020 and have several resolutions varying from 360×160 to 1862×1008 pixels – depending on the mobile device used to capture the image. In total, the dataset has images captured from 196 different mobile devices. Each subject captured their images using the same device model. This dataset's main intra- and inter-class variability are caused by lighting variation, occlusion, specular reflection, blur, motion blur, eyeglasses, off-angle, eye-gaze, makeup, and facial expression. The authors manually annotated the eye corner of all images with 4 points (inside and outside eye corners) and used it to normalize the periocular region regarding scale and rotation. All the original and cropped periocular images, eye-corner annotations, and experimental protocol files are publicly available for the research community (upon request). The paper contains information about images' distributions by gender, age, resolution, and other experiments' details and benchmarks. Source: A new periocular dataset collected by mobile devices in unconstrained scenarios

Urban Hyperspectral Image

Urban is one of the most widely used hyperspectral data used in the hyperspectral unmixing study. There are 307x307 pixels, each of which corresponds to a 2x2 m2 area. In this image, there are 210 wavelengths ranging from 400 nm to 2500 nm, resulting in a spectral resolution of 10 nm. After the channels 1-4, 76, 87, 101-111, 136-153 and 198-210 are removed (due to dense water vapor and atmospheric effects), we remain 162 channels (this is a common preprocess for hyperspectral unmixing analyses). There are three versions of ground truth, which contain 4, 5 and 6 endmembers respectively, which are introduced in the ground truth. Linda S. Kalman and Edward M. Bassett III "Classification and material identification in an urban environment using HYDICE hyperspectral data", Proc. SPIE 3118, Imaging Spectrometry III, (31 October 1997); https://doi.org/10.1117/12.283843 Hosted at: - https://rslab.ut.ac.ir/data - http://lesun.weebly.com/hyperspectral-data-set.html - https://erdc-library.erdc.dren.mil/jspui/handle/11681/2925

5,011 Images – Human Frontal face Data (Male)

Description: 5,011 Images – Human Frontal face Data (Male). The data diversity includes multiple scenes, multiple ages and multiple races. This dataset includes 2,004 Caucasians , 3,007 Asians. This dataset can be used for tasks such as face detection, race detection, age detection, beard category classification. Data size: 5,011 people, one image per person Race distribution: 2,004 Caucasians , 3,007 Asians

AnimeCeleb

We present a novel Animation CelebHeads dataset (AnimeCeleb) to address an animation head reenactment. Different from previous animation head datasets, we utilize 3D animation models as the controllable image samplers, which can provide a large amount of head images with their corresponding detailed pose annotations. To facilitate a data creation process, we build a semi-automatic pipeline leveraging an open 3D computer graphics software with a developed annotation system. After training with the AnimeCeleb, recent head reenactment models produce high-quality animation head reenactment results, which are not achievable with existing datasets. Furthermore, motivated by metaverse application, we propose a novel pose mapping method and architecture to tackle a cross-domain head reenactment task. During inference, a user can easily transfer one's motion to an arbitrary animation head. Experiments demonstrate the usefulness of the AnimeCeleb to train animation head reenactment models, and the superiority of our cross-domain head reenactment model compared to state-of-the-art methods. Our dataset and code are available at https://github.com/kangyeolk/AnimeCeleb.

Bone Age

At RSNA 2017 there was a contest to correctly identify the age of a child from an X-ray of their hand.

Celeb-HQ Face Gender Recognition Dataset

Celeb-HQ Face Gender Recognition Dataset * This dataset is curated for the face gender classification task. * The dataset contains 30,000 images. * There are 23,999 train images. * There are 6,001 test images. * The whole face images are divided into two classes. * There are 11,057 male images. * There are 18,943 female images.

Celeb-HQ Facial Identity Recognition Dataset

Celeb-HQ Facial Identity Recognition Dataset * This dataset is curated for the facial identity classification task. * There are 307 identities (celebrities). * Each identity has 15 or more images. * The dataset contains 5,478 images. * There are 4,263 training images. * There are 1,215 test images.

DARK FACE

DARK FACE dataset provides 6,000 real-world low light images captured during the nighttime, at teaching buildings, streets, bridges, overpasses, parks etc., all labeled with bounding boxes for of human face, as the main training and/or validation sets. We also provide 9,000 unlabeled low-light images collected from the same setting. Additionally, we provided a unique set of 789 paired low-light/normal-light images captured in controllable real lighting conditions (but unnecessarily containing faces), which can be used as parts of the training data at the participants' discretization. There will be a hold-out testing set of 4,000 low-light images, with human face bounding boxes annotated.

EDFace-Celeb-1M

EDFace-Celeb-1M is a public Ethnically Diverse Face dataset which is used to benchmark the task of face hallucination. The dataset includes 1.7 million photos that cover different countries, with balanced race composition.

FaMoS

FaMoS is a dynamic 3D head dataset from 95 subjects, each performing 28 motion sequences. The sequences comprise of six prototypical expressions (i.e., Anger, Disgust, Fear, Happiness, Sadness, and Surprise), two head rotations (left/right and up/down), and diverse facial motions, including extreme and asymmetric expressions. Each sequence is recorded at 60 fps. In total, FaMoS contains around 600K 3D head meshes (i.e., ~225 frames per sequence). For each frame, registrations in FLAME meshes are publicly available.

FDDB-360

A 360-degree fisheye-like version of the popular FDDB face detection dataset. Source: FDDB-360: Face Detection in 360-degree Fisheye Images

HDA Facial Tattoo and Painting Database

The Hochschule Darmstadt (HDA) facial tattoo and paintings database contains 500 pairs of facial images of individuals with and without facial tattoos or paintings. The database was collected from multiple online sources.

KANFace

KANFace consists of 40K still images and 44K sequences (14.5M video frames in total) captured in unconstrained, real-world conditions from 1,045 subjects. The dataset is manually annotated in terms of identity, exact age, gender and kinship. Source: KANFace Dataset

MFW+ (U-M)

MFW+ is a benchmark dataset for masked face recognition and an extended version of MFW. The original MFW, published as a benchmark for masked face recognition, is composed of 300 IDs and 3,000 images. However, with two duplicate IDs found in MFW, the dataset actually contains 298 unique IDs and 2,980 images. To evaluate models under various mask conditions and environments, we manually gathered additional data from the web. The refined and extended MFW, which we named MFW+, contains 606 IDs, 2,911 unmasked face images, and 2,838 masked face images. Paper: https://bmvc2022.mpi-inf.mpg.de/0723.pdf

PetFace

PetFace is a large-scale animal face re-identification dataset that includes 257,484 unique individuals across 13 families and 319 breeds. PetFace has fine-grained annotation (sex, breeds, color, and patterns).

RLU

RL Unplugged is suite of benchmarks for offline reinforcement learning. The RL Unplugged is designed around the following considerations: to facilitate ease of use, we provide the datasets with a unified API which makes it easy for the practitioner to work with all data in the suite once a general pipeline has been established. This is a dataset accompanying the paper RL Unplugged: Benchmarks for Offline Reinforcement Learning. In this suite of benchmarks, we try to focus on the following problems: High dimensional action spaces, for example the locomotion humanoid domains, we have 56 dimensional actions. High dimensional observations. Partial observability, observations have egocentric vision. Difficulty of exploration, using states of the art algorithms and imitation to generate data for difficult environments. Real world challenges.

SKSF-A

SKSF-A consists of seven distinct styles drawn by professional artists. SKSF-A contains 134 identities and corresponding sketches, making a total of 938 face-sketch pairs. SKSF-A is introduced in StyleSketch, Eurographics 2024. https://kwanyun.github.io/stylesketch_project/

SuHiFiMask

SuHiFiMask (Surveillance High-Fidelity Mask) extends FAS to real surveillance scenes rather than mimicking low-resolution images and surveillance environments. It contains 10,195 videos from 101 subjects of different age groups, which are collected by 7 mainstream surveillance cameras. Source: Surveillance Face Presentation Attack Detection Challenge

SWAX

Comprised of real human and wax figure images and videos that endorse the problem of face spoofing detection. The dataset consists of more than 1800 face images and 110 videos of 55 people/waxworks, arranged in training, validation and test sets with a large range in expression, illumination and pose variations. Source: The SWAX Benchmark: Attacking Biometric Systems with Wax Figures

Udacity

The Udacity dataset is mainly composed of video frames taken from urban roads. It provides a total number of 404,916 video frames for training and 5,614 video frames for testing. This dataset is challenging due to severe lighting changes, sharp road curves and busy traffic. Source: Learning to Steer by Mimicking Features from Heterogeneous Auxiliary Networks Image Source: https://www.researchgate.net/figure/Sample-from-the-Udacity-dataset-with-the-original-ground-truth-bounding-boxes-Note-that_fig3_345652980

VideoForensicsHQ

VideoForensicsHQ is a benchmark dataset for face video forgery detection, providing high quality visual manipulations. It is one of the first face video manipulation benchmark sets that also contains audio and thus complements existing datasets along a new challenging dimension. VideoForensicsHQ shows manipulations at much higher video quality and resolution, and shows manipulations that are provably much harder to detect by humans than videos in other datasets. VideoForensicsHQ contains 1,737 videos of speaking faces (44% male, 56% female), with 8 different emotions, most of them of “HD” resolution. The videos amount to 1,666,816 frames. Source: VideoForensicsHQ: Detecting High-quality Manipulated Face Videos

23 Pairs of Identical Twins Face Image Data

Description: 23 Pairs of Identical Twins Face Image Data. The collecting scenes includes indoor and outdoor scenes. The subjects are Chinese males and females. The data diversity inlcudes multiple face angles, multiple face postures, close-up of eyes, multiple light conditions and multiple age groups. This dataset can be used for tasks such as twins' face recognition. Data size: 23 pairs, each person in a pair of identical twins has 40 images (20 indoor images, 20 outdoor images) Population distribution: race distribution: Asian (Chinese); gender distribution: male 9 pairs, female 14 pairs; age distribution: 12 pairs under 18 years old, 10 pairs aged from 18 to 40, 1 pairs over 40 years old

A View From Somewhere (AVFS)

A View From Somewhere (AVFS)—a dataset of 638,180 face similarity judgments over 4,921 faces. Each judgment corresponds to the odd-one-out (i.e., least similar) face in a triplet of faces and is accompanied by both the identifier and demographic attributes of the annotator who made the judgment.

Age and Gender

EEG signals from 60 users have been recorded whose age range lies between 6 and 55 years. Among all, there were 25 females and 35 male users. In general, all the participants were either school children or belonged to the socioeconomic cross section of the population with no medical history. The EEG recordings were acquired from all 14 electrodes operating at a sampling rate of 128 Hz. During recording, the participants were asked to comfortably sit on the chair with clear thoughts and a relaxed state.

BN-AuthProf

Although research on author profiling has quite progressed in abundant resources languages, it is still infancy for limited resources languages such as Bengali. This repository contains our Bangla Author Profiling Dataset (BN-AuthProf). The primary objective is to introduce and benchmark the performance of machine learning approaches on Age and Gender Classification tasks from the social media status of people. BN-AuthProf dataset consists of 300 anonymized authors and 30,131 manually curated Facebook posts in Bangla, tagged with age and gender information.

CANDOR Corpus

The CANDOR corpus is a large, novel, multimodal corpus of 1,656 recorded conversations in spoken English. This 7+ million word, 850 hour corpus totals over 1TB of audio, video, and transcripts, with moment-to-moment measures of vocal, facial, and semantic expression, along with an extensive survey of speaker post conversation reflections.

ChildCIdb

A large-scale, first-of-its-kind database aimed at generating a better understanding of the way children interact with mobile devices during their development process. ChildCIdbv1 comprises data collected from 438 children, from 18 months to 8 years old, encompassing the first three development stages of Piaget's theory. Data collected spans interaction with screens using both finger and pen stylus, information regarding the previous experience of the child with mobile devices, the child’s grade level, and whether attention-deficit/hyperactivity disorder (ADHD) is present. Use cases: Child age detection based on device interaction.

Curated AFD

The Curated AFD dataset is a curated version of the Asian Face Dataset (AFD) for face recognition research. The original AFD dataset has been curated to remove wrong identity labels, duplicate images and duplicate subjects. Source: https://arxiv.org/abs/2004.03074

Extended YouTube Faces (E-YTF)

The proposed Extended-YouTube Faces (E-YTF) is an extension of the famous YouTube Faces (YTF) dataset and is specifically designed to further push the challenges of face recognition by addressing the problem of open-set face identification from heterogeneous data i.e. still images vs video.

Facial Skeletal angles

Facial Skeletal Angles (Glabella and Maxilla Angle and Length and Width of Piriformis)

FER2013 Blendshapes

Tables of the blendshapes from a group of the images of the FER2013 dataset, generated using MediaPipe library, based on the ARKit face blendshapes. with classes of the images in a separate column, describing the categories Happy, Unknown, Sad.

FLUXSynID

The FLUXSynID Dataset comprises 14,889 high-resolution synthetic face identities, each uniquely represented with paired images: document-style images and trusted live-capture images. The document-style images are designed to simulate ID photos, while the live-capture images are generated to reflect real-world conditions, including variations in lighting, facial expressions, and camera angles. This dual representation facilitates robust benchmarking for face recognition, morphing attack detection, and cross-domain identity verification tasks. FLUXSynID leverages cutting-edge generative models to ensure high fidelity and diversity across identities, enabling comprehensive analysis of biometric systems under varied conditions. For more information and to generate your own custom datasets using the FLUXSynID framework, visit the FLUXSynID GitHub repository.

IMFW

Indian Masked faces in the wild Database is collected into three sets:(i) Indian Celebrity, (ii) Instagram and (iii) Indian Crowd. The Indian Celebrity contains 40 Indian celebrities with 435 images, including Bollywood actors/actresses, television stars, sports personalities, and politicians. The Instagram set contains 377 images of 40 subjects downloaded from Instagram. We collected masked and non-masked images of Indian people with a public profile. The Indian Crowd set is collected from the common people who volunteered to contribute to the dataset. This set contains 120 subjects with 562 images. All the Images are collected in both constrained and unconstrained environments with variation in pose, illumination, background and masks worn by the people.

출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.